Skip to main content
. 2017 Jan 25;12(1):e0169754. doi: 10.1371/journal.pone.0169754

Table 2. RMSE for predicted abundance gradients.

RMSE has been calculated between the predicted gradients and the target gradients, both in silico and in vitro, having used LDA and a Random Forest classifier.

Comb. 1 Comb. 2 Comb. 3
RMSE LDA in silico 0.29 0.0060 0.10
RMSE LDA in vitro 0.51 0.036 0.096
RMSE RF in silico 0.21 0.0036 0.022
RMSE RF in vitro 0.48 0.036 0.032